Migrate Airflow to ZenML

This skill translates Apache Airflow DAGs into idiomatic ZenML pipelines. It handles the full migration workflow: analyzing Airflow code, classifying each pattern, translating what maps cleanly, flagging what needs redesign, and producing a working ZenML project.

How migration works at a high level

Airflow and ZenML look similar on the surface — DAG maps to pipeline, operator maps to step, XCom maps to artifact — but their execution models are fundamentally different. Airflow is built around a scheduler-backed, database-persisted task-instance state machine. ZenML is built around artifact lineage, stack-driven infrastructure abstraction, and Python-first pipeline composition.

This means migration is not a rename-the-primitives exercise. Some patterns translate directly, some need approximation, and some require genuine redesign. The skill's job is to be honest about which is which.

The three mapping types

Every Airflow concept falls into one of these categories: